AI security

Prompt injection is product security

The attacker does not need shell access if your assistant will politely leak the thing it was told to protect.

~5 minAI
scroll
01 · the shape

The prompt is not a perimeter

A system prompt feels private because users cannot see it. That does not make it a boundary. The model is still reading untrusted text and trusted instructions in the same mental room, and attackers know how to make those voices collide.

chat · hostile content

Untrusted text can arrive through docs, tickets, pages, and emails.

guardrails armed

Hit the button and watch the hidden AI surface area light up.

02 · the move

Separate instructions from evidence

Treat retrieved content like user input, even when it came from your own database. Keep privileged tools behind explicit policy checks, quote untrusted text as evidence, and make the model ask for permission before crossing a sensitive boundary.

guards/tool-policy.js

The model can request a tool. Policy decides whether it gets one.

03 · the check

Test the ugly inputs

A safe AI feature needs hostile examples in the test set: jailbreaks, fake system messages, poisoned search results, and requests that try to turn summaries into actions.

AI launch switchboard
sloppy
shippable0
fragile: the model is still freelancing without enough checks

Flip the switches and watch the story turn into a tiny operating model.

0places to guard: input, retrieval, tools
0policy layer outside the model
0secrets placed in prompts
04 · the takeaway

Do not ask the model to be the firewall

The model can help reason about intent, but it should not be the only thing standing between a clever sentence and a privileged action. Put boring controls around the exciting box.

source trail